Vertex CLI

A modern CLI tool for interacting with multiple LLMs using LangChain.

Features

• Unified interface for multiple LLM providers (Google Gemini, OpenAI, Anthropic Claude)
• Clean architecture following SOLID principles
• Conversation history management
• Shell command debugging
• Easy model switching

Installation

pip install Vertex-CLI

Gemini works out of the box. OpenAI and Anthropic are optional extras, so their packages are only downloaded if you ask for them:

pip install "Vertex-CLI[openai]"     # OpenAI models
pip install "Vertex-CLI[anthropic]"  # Anthropic Claude models

To work on Vertex-CLI itself, install from a clone instead:

git clone https://github.com/prathamhole14/Vertex-CLI.git
cd Vertex-CLI
pip install -e ".[openai,anthropic]"

Quick Start

# Initialize configuration
tex --setup

# Configure a model (provider auto-detected from model name)
tex config gemini-3.6-flash YOUR_API_KEY
tex config gpt-4 YOUR_OPENAI_KEY
tex config claude-opus-5 YOUR_ANTHROPIC_KEY

# Or specify provider explicitly
tex config my-model YOUR_KEY --provider google

# Select active model
tex select gemini-3.6-flash

# Use it
tex "explain quantum computing"

Commands

tex "your question"                  # Ask a question
tex chat <text>                      # Chat mode
tex debug                            # Debug shell commands
tex config <model> <api_key>         # Configure model (auto-detects provider)
tex config <model> <key> --provider <provider>  # Specify provider explicitly
tex list                             # List models
tex select <model>                   # Select active model
tex remove <model>                   # Remove model

Supported Providers

• Google Gemini: google or gemini — installed by default
• OpenAI: openai — needs pip install 'Vertex-CLI[openai]'
• Anthropic: anthropic or claude — needs pip install 'Vertex-CLI[anthropic]'

Configuration

Models are configured in ~/.config/ai_model_manager/models_config.json

Example configuration:

{
  "selected_model": "gemini-flash-latest",
  "models": {
    "gemini-flash-latest": {
      "provider": "google",
      "api_key": "your-api-key",
      "temperature": 0.7
    }
  }
}

Documentation

• Architecture - Technical architecture

Requirements

• Python 3.10+
• LangChain and provider packages

Usage

Once installed and configured, you can start chatting or debugging commands:

Chat with the LLM

You can either use the chat subcommand or omit it entirely:

# Explicit subcommand
tex chat "Tell me about the solar system"

# Shortcut form (no subcommand)
tex "Tell me about the solar system"

Replace the quoted string with any query you'd like.

🔗 Complete CLI Documentation: CLI Commands

Debugging Mode (Beta Feature)

Debugging is currently in beta but can analyze recent shell commands to identify issues.

Debug the Last 3 Commands (default)

tex debug

Specify the Number of Commands to Debug

tex debug -n 5

Add a Custom Debugging Message

tex debug -n 5 -p "Explain why \`git commit\` failed"

Contributing

Contributions are welcome! Follow these steps to contribute:

1. Fork the repository

2. Create a new branch:

   git checkout -b feature/your-feature-name
   

3. Commit your changes:

   git commit -m "Add your feature description"
   

4. Push your branch:

   git push origin feature/your-feature-name
   

5. Open a pull request

🔗 Contributor Guide: How to Contribute

License

This project is licensed under the MIT License. See the LICENSE file for more details.

Support

If you encounter any issues, open an issue on the GitHub repository.